Abstract
The textile industry creates one-fifth of the world's industrial water pollution, thus, the electrocoagulation (EC) process was proposed and investigated as an alternative eco-friendly treatment for water reuse. This study aimed to assess the removal efficiency of the Reactive Black 5 (RB5) from synthetic textile effluent by EC. Key operating parameters on EC process efficiency were optimized using response surface methodology (RSM). The main independent variables studied were the current density, EC time, concentration, and pH while the RB5 dye removal was studied as the dependent variable. The range of the studied parameters affecting the EC process ranged from 10 to 60 mA/cm2, 5 to 30 min, 4 to 10 pH and 10 to 40 ppm for current density, EC time, pH, and RB5 concentration, respectively. The optimal operating parameters turned out to be 5.5 pH, 47.5 mA/cm2, 23.75 min, 17.5 ppm, and the predicted RB5 dye removal was 96.33%. The experimental dye removal with optimum operating conditions was in good agreement with the predicted removal efficiency. Therefore, the experiment results revealed the high potential of the EC process to effectively treat textile industry effluents and the RB5 dye removal was successfully optimized using Response Surface Methodology (RSM).
HIGHLIGHTS
The removal of Reactive Black 5 from simulated textile effluent by electrocoagulation process was investigated.
RB5 dye removal process was successfully optimized using RSM.
The experimental dye removal with optimum operating conditions was in good agreement with the predicted removal efficiency.
The experiment results proved that the electrocoagulation process is a potential eco-friendly process to treat textile industry effluents.
INTRODUCTION
According to the literature, the EC process turned out to be very efficient for the abatement of a wide range of pollutants (Emamjomeh & Sivakumar 2009), thus, this process may act as an excellent technology for textile wastewater treatment. The dye removal from wastewater is a huge challenge for most conventional treatment (Zaroual et al. 2006; Kobya et al. 2016; Meddah et al. 2021). Some dyes are toxic and are not easily biodegraded by biological processes (Roriz et al. 2009; Kabdaşlı & Tünay 2012; Meddah et al. 2021). Chemical coagulation–flocculation can be used for dye removal, but it is ineffective for decolorizing textile effluents. Also, the addition of chemicals into water presents its drawback (Zaroual et al. 2006; Yousefi et al. 2022). EC has been considered for the treatment of dye streams from textile industries wastewater and it does not need adding chemicals for the process which makes it an eco-friendly process (Khandegar & Saroha 2014). The EC process combines physical and chemical mechanisms with many electrochemical phenomena involved.
The purely physical enmeshment of dissolved substances during hydroxide precipitation, adsorption, and complexation.
The electro-oxidation on the anode or the electro-reduction on the cathode of electro-active ions or molecules.
The direct adsorption of pollutants on the electrodes.
As RB5 is the main azo dye widely used in the textile industry and the research on RB5 dye effluent treatment by the EC process is almost non-existent in literature, the current study aims at investigating and validating the potential applicability of the EC process as a mature technology to treat textile effluent for its reuse. The RB5 removal efficiency by the EC process is assessed in batch mode with aluminum electrodes and operating parameters are optimized using surface response methodology. Interaction between the different independent factors such as the current density, EC time, concentration, and initial pH and their influence on the dependent factor which is the RB5 dye removal efficiency are assessed.
MATERIALS AND METHODS
Experimental set up and procedure
The mass of the floc formed is proportional to the mass of aluminum released by electrodes and it changes over time (Daneshvar et al. 2007). This is clearly proven by Faraday's law using the Faradaic yield constant. It is calculated by comparing the experimental mass loss of aluminum electrodes during the EC process with the theoretical amount of aluminum dissolution, based on Faraday's law.
Statistical analysis and design of experiment
It means that is obtained due to the four (4) factors used for the study. Design Expert 13 software was used for designing the experiment and analysis of data.
RESULTS AND DISCUSSION
RSM results
The most EC response influential independent factors were chosen as shown in Table 1 with their corresponding coded levels. The four parameters were used in coded form based on minimum, maximum, and points between minimum and maximum. The factors with their respective alpha () and () values are: Current density (10–60 mA/cm2), EC time (5–30 min), concentration (10–40 ppm), and pH (4–10). Therefore, 31 experimental runs with corresponding experimental and predicted dye percent removal are presented in Table 2.
Parameters . | Code . | Coded factors level . | ||||
---|---|---|---|---|---|---|
− 2 . | − 1 . | 0 . | 1 . | 2 . | ||
Current density (mA/cm2) | A | 10 | 22.5 | 35 | 47.5 | 60 |
EC time (min) | B | 5 | 11.25 | 17.5 | 23.75 | 30 |
Concentration (ppm) | C | 10 | 17.5 | 25 | 32.5 | 40 |
pH | D | 4 | 5.5 | 7 | 8.5 | 10 |
Parameters . | Code . | Coded factors level . | ||||
---|---|---|---|---|---|---|
− 2 . | − 1 . | 0 . | 1 . | 2 . | ||
Current density (mA/cm2) | A | 10 | 22.5 | 35 | 47.5 | 60 |
EC time (min) | B | 5 | 11.25 | 17.5 | 23.75 | 30 |
Concentration (ppm) | C | 10 | 17.5 | 25 | 32.5 | 40 |
pH | D | 4 | 5.5 | 7 | 8.5 | 10 |
Run order . | A (current density [mA/Cm2]) . | B (electrolysis time[min]) . | C (concentration [ppm]) . | D (pH) . | % Actual RB5 . | % Predicted RB5 . |
---|---|---|---|---|---|---|
1 | 0 (35) | 0 (17.5) | 0 (25) | 0 (7) | 61.84 | 65.28 |
2 | 0 (35) | 0 (17.5) | 0 (25) | −2 (4) | 65.09 | 63.20 |
3 | 2 (60) | 0 (17.5) | 0 (25) | 0 (7) | 95.20 | 86.49 |
4 | 0 (35) | 0 (17.5) | 0 (25) | 2 (10) | 35.40 | 41.97 |
5 | 0 (35) | 0 (17.5) | 0 (25) | 0 (7) | 64.62 | 65.28 |
6 | 1 (47.5) | −1 (11.25) | −1 (17.5) | 1 (8.5) | 64.16 | 56.48 |
7 | 0 (35) | 0 (17.5) | −2 (10) | 0 (7) | 64.67 | 72.66 |
8 | −1 (22.5) | 1 (23.75) | −1 (17.5) | −1 (5.5) | 71.02 | 68.74 |
9 | −1 (22.5) | −1 (11.25) | 1 (32.5) | 1 (8.5) | 35.27 | 34.27 |
10 | −1 (22.5) | −1 (11.25) | −1 (17.5) | −1 (5.5) | 53.07 | 52.26 |
11 | 0 (35) | 2 (30) | 0 (25) | 0 (7) | 93.01 | 88.14 |
12 | 1 (47.5) | 1 (23.75) | 1 (32.5) | 1 (8.5) | 78.16 | 78.33 |
13 | 0 (35) | −2 (5) | 0 (25) | 0 (7) | 38.33 | 42.42 |
14 | 0 (35) | 0 (17.5) | 0 (25) | 0 (7) | 62.86 | 65.28 |
15 | 1 (47.5) | 1 (23.75) | −1 (17.5) | −1 (5.5) | 94.23 | 96.33 |
16 | 0 (35) | 0 (17.5) | 0 (25) | 0 (7) | 63.56 | 65.28 |
17 | 0 (35) | 0 (17.5) | 0 (25) | 0 (7) | 65.02 | 65.28 |
18 | −1 (22.5) | −1 (11.25) | −1 (17.5) | 1 (8.5) | 49.63 | 41.65 |
19 | −1 (22.5) | 1 (23.75) | 1 (32.5) | −1 (5.5) | 59.65 | 61.35 |
20 | 0 (35) | 0 (17.5) | 0 (25) | 0 (7) | 64.92 | 65.28 |
21 | 1 (47.5) | −1 (11.25) | 1 (32.5) | 1 (8.5) | 43.55 | 49.09 |
22 | 1 (47.5) | −1 (11.25) | 1 (32.5) | −1 (5.5) | 61.11 | 59.70 |
23 | 0 (35) | 0 (17.5) | 0 (25) | 0 (7) | 62.54 | 65.28 |
24 | 1 (47.5) | −1 (11.25) | −1 (17.5) | −1 (5.5) | 64.82 | 67.09 |
25 | −2 (10) | 0 (17.5) | 0 (25) | 0 (7) | 37.02 | 44.07 |
26 | 0 (35) | 0 (17.5) | 2 (40) | 0 (7) | 61.00 | 57.90 |
27 | −1 (22.5) | 1 (23.75) | 1 (32.5) | 1 (8.5) | 51.54 | 50.74 |
28 | 1 (47.5) | 1 (23.75) | −1 (17.5) | 1 (8.5) | 85.10 | 85.72 |
29 | −1 (22.5) | 1 (23.75) | −1 (17.5) | 1 (8.5) | 64.83 | 58.13 |
30 | −1 (22.5) | −1 (11.25) | 1 (32.5) | −1 (5.5) | 52.15 | 44.88 |
31 | 1 (47.5) | 1 (23.75) | 1 (32.5) | −1 (5.5) | 84.14 | 88.95 |
Run order . | A (current density [mA/Cm2]) . | B (electrolysis time[min]) . | C (concentration [ppm]) . | D (pH) . | % Actual RB5 . | % Predicted RB5 . |
---|---|---|---|---|---|---|
1 | 0 (35) | 0 (17.5) | 0 (25) | 0 (7) | 61.84 | 65.28 |
2 | 0 (35) | 0 (17.5) | 0 (25) | −2 (4) | 65.09 | 63.20 |
3 | 2 (60) | 0 (17.5) | 0 (25) | 0 (7) | 95.20 | 86.49 |
4 | 0 (35) | 0 (17.5) | 0 (25) | 2 (10) | 35.40 | 41.97 |
5 | 0 (35) | 0 (17.5) | 0 (25) | 0 (7) | 64.62 | 65.28 |
6 | 1 (47.5) | −1 (11.25) | −1 (17.5) | 1 (8.5) | 64.16 | 56.48 |
7 | 0 (35) | 0 (17.5) | −2 (10) | 0 (7) | 64.67 | 72.66 |
8 | −1 (22.5) | 1 (23.75) | −1 (17.5) | −1 (5.5) | 71.02 | 68.74 |
9 | −1 (22.5) | −1 (11.25) | 1 (32.5) | 1 (8.5) | 35.27 | 34.27 |
10 | −1 (22.5) | −1 (11.25) | −1 (17.5) | −1 (5.5) | 53.07 | 52.26 |
11 | 0 (35) | 2 (30) | 0 (25) | 0 (7) | 93.01 | 88.14 |
12 | 1 (47.5) | 1 (23.75) | 1 (32.5) | 1 (8.5) | 78.16 | 78.33 |
13 | 0 (35) | −2 (5) | 0 (25) | 0 (7) | 38.33 | 42.42 |
14 | 0 (35) | 0 (17.5) | 0 (25) | 0 (7) | 62.86 | 65.28 |
15 | 1 (47.5) | 1 (23.75) | −1 (17.5) | −1 (5.5) | 94.23 | 96.33 |
16 | 0 (35) | 0 (17.5) | 0 (25) | 0 (7) | 63.56 | 65.28 |
17 | 0 (35) | 0 (17.5) | 0 (25) | 0 (7) | 65.02 | 65.28 |
18 | −1 (22.5) | −1 (11.25) | −1 (17.5) | 1 (8.5) | 49.63 | 41.65 |
19 | −1 (22.5) | 1 (23.75) | 1 (32.5) | −1 (5.5) | 59.65 | 61.35 |
20 | 0 (35) | 0 (17.5) | 0 (25) | 0 (7) | 64.92 | 65.28 |
21 | 1 (47.5) | −1 (11.25) | 1 (32.5) | 1 (8.5) | 43.55 | 49.09 |
22 | 1 (47.5) | −1 (11.25) | 1 (32.5) | −1 (5.5) | 61.11 | 59.70 |
23 | 0 (35) | 0 (17.5) | 0 (25) | 0 (7) | 62.54 | 65.28 |
24 | 1 (47.5) | −1 (11.25) | −1 (17.5) | −1 (5.5) | 64.82 | 67.09 |
25 | −2 (10) | 0 (17.5) | 0 (25) | 0 (7) | 37.02 | 44.07 |
26 | 0 (35) | 0 (17.5) | 2 (40) | 0 (7) | 61.00 | 57.90 |
27 | −1 (22.5) | 1 (23.75) | 1 (32.5) | 1 (8.5) | 51.54 | 50.74 |
28 | 1 (47.5) | 1 (23.75) | −1 (17.5) | 1 (8.5) | 85.10 | 85.72 |
29 | −1 (22.5) | 1 (23.75) | −1 (17.5) | 1 (8.5) | 64.83 | 58.13 |
30 | −1 (22.5) | −1 (11.25) | 1 (32.5) | −1 (5.5) | 52.15 | 44.88 |
31 | 1 (47.5) | 1 (23.75) | 1 (32.5) | −1 (5.5) | 84.14 | 88.95 |
The experimental results were analyzed through CCD in RSM to obtain a response. The actual RB5 percentage (%RB5) removal represented the measured response data for a given experimental run while the predicted RB5 percentage (%RB5) removal was deduced from the second polynomial model generated from RSM.
Analysis of variance (ANOVA)
The regression equation (second-order polynomial) model was obtained as the relationship between factors and response. The equation is expressed as follows:
The results of the analysis of variance for the reduced model in Table 3 were established with new regression model Equation (10) with the following parameters: EC time, current density, concentration and pH. All of them are significant in their linear terms. The p-value acts as the reference to ensure the significance of each parameter (Razieh et al. 2020; Meddah et al. 2021). Due to that, p-value was estimated at a 5% level, by considering the results obtained at linearity are even significant at a 1% level. The pH quadratic factor, the current density and EC time interaction factor have shown the p-values of 0.0023 and 0.018, which is less than p-value of 0.05. They are significant at a 95% confidence level. The R2 value of 92.26%, the adjusted R2 (R2adj) of 90.33% and the predicted R2 (R2pred) of 84.75% have shown the values that are closer and they present a satisfactory adjustment of the quadratic model to the experimental results. The difference between R2pred and R2adj should be less than 0.20 in order to confirm that the results or model are reliable (Mook et al. 2017; Razieh et al. 2020). Therefore, R2pred is in reasonable agreement with the R2adj which indicates the validity and suitability of the obtained model to fit the RB5 removal efficiency data.
Source . | Sum of squares . | df . | Mean square . | F-value . | p-value . |
---|---|---|---|---|---|
Model | 7,295.74 | 6 | 1,215.96 | 47.69 | < 0.0001 |
A – current density | 2,698.61 | 1 | 2,698.61 | 105.84 | < 0.0001 |
B – EC time | 3,135.00 | 1 | 3,135.00 | 122.95 | < 0.0001 |
C – concentration | 327.12 | 1 | 327.12 | 12.83 | 0.0015 |
D – pH | 675.73 | 1 | 675.73 | 26.50 | < 0.0001 |
AB | 162.99 | 1 | 162.99 | 6.39 | 0.0185 |
D² | 296.29 | 1 | 296.29 | 11.62 | 0.0023 |
Residual | 611.94 | 24 | 25.50 | ||
Lack of fit | 602.36 | 18 | 33.46 | 20.96 | 0.0006 |
Pure error | 9.58 | 6 | 1.60 | ||
Cor. total | 7,907.68 | 30 |
Source . | Sum of squares . | df . | Mean square . | F-value . | p-value . |
---|---|---|---|---|---|
Model | 7,295.74 | 6 | 1,215.96 | 47.69 | < 0.0001 |
A – current density | 2,698.61 | 1 | 2,698.61 | 105.84 | < 0.0001 |
B – EC time | 3,135.00 | 1 | 3,135.00 | 122.95 | < 0.0001 |
C – concentration | 327.12 | 1 | 327.12 | 12.83 | 0.0015 |
D – pH | 675.73 | 1 | 675.73 | 26.50 | < 0.0001 |
AB | 162.99 | 1 | 162.99 | 6.39 | 0.0185 |
D² | 296.29 | 1 | 296.29 | 11.62 | 0.0023 |
Residual | 611.94 | 24 | 25.50 | ||
Lack of fit | 602.36 | 18 | 33.46 | 20.96 | 0.0006 |
Pure error | 9.58 | 6 | 1.60 | ||
Cor. total | 7,907.68 | 30 |
Effect of factors interaction on RB5 dye removal
Interaction between current density and EC time
The EC time and the current density interaction as shown in Figure 8(a) demonstrated a great influence on RB5 dye removal efficiency. The latter increased with increasing time and current density while initial concentration and pH were set constant at midpoints 25 ppm and 7, respectively. From the current density of 47.5 mA/cm2 and treatment time of 23.75 min to 60 mA/cm2 and 30 min, the surface plot predicted the RB5 removal efficiency that increased from 80 to 100%, respectively. An increase in current density and EC time resulted in an increasing RB5 removal efficiency. The current density which is defined as the current over the active electrode surface area determines the amount of coagulant electrochemically released in the EC reactor and EC (electrolysis) time in return increases the amount of coagulant according to Faraday's law. Thus, increasing the current density and EC time at the same time results in increasing considerably the RB5 removal efficiency (Kobya et al. 2016; Hakizimana et al. 2017). Furthermore, on the one hand, the current density is also known to control the production and size of hydrogen bubbles at the cathode which can enhance the mass transfer between hydroxide anions and metal cations resulting in the efficient formation of aluminum hydroxides (coagulant) (Daneshvar et al. 2006). On the other hand, these hydrogen bubbles produced at the cathodes encounter the flocs that are attached to the bubble surface which brings them to the top of the EC reactor. This constituted the main liquid-solid phase separation associated with the EC, known as the electroflotation process (Hakizimana et al. 2017). The EC time increased the amount of coagulant and enhanced the contact between the coagulant and the RB5 dye which resulted in increasing RB5 dye removal efficiency (Mook et al. 2017).
Interaction between current density and RB5 dye concentration
The effect of current density combined with that of RB5 concentration is shown in Figure 8(b). A better RB5 removal is obtained at high current density with slightly low concentration. At the current density of 23.75 min and concentration of 17.5 ppm the RB5 removal efficiency has been expected to reach between 70 and 80%. The lowest percentage was observed at high concentration and low current density with 40% removal. The increase of concentration requires a huge amount of the coagulant in the EC reactor, which involves increasing the current density. Increasing the current density augments the metal cations, hydroxyl anions and hydrogen bubbles electrochemically produced during the EC process. The efficient formation of Al(OH)3 enhances a large surface area to remove the dye (Mollah et al. 2010) and considerable hydrogen bubbles production increases the contact between Al(OH)3 coagulants and RB5 dye. Thus, the removal efficiency of RB5 increases with decreasing moderate concentration and increasing current density (Mook et al. 2017).
Interaction between pH and current density
Interaction between RB5 concentration and EC time
The high concentration and the short EC time in Figure 8(d) shows that there is a remarkable decrease in RB5 removal efficiency. It is reported by Chang et al. (2010), that the increase in concentration requires a considerable amount of coagulant in the EC process, which entails EC long time at constant current density. In Figure 8(d), EC time set at 23.75 min and the concentration within the range between 13.75–17.5 ppm, the RB5 removal efficiency reached between 80 and 90%. Therefore, a highly concentrated dye requires an electrochemical dissolution that needs to be generated for a long time to trap the dye. The slightly low concentration with the long electrolysis time led to the high RB5 removal efficiency.
Interaction between pH and EC time
The Initial pH and the EC time in Figure 8(e) influenced greatly the performance of the EC process. Between pH of 4 to 7.6, the RB5 removal efficiency increased from 25% at 5 min to 90% at 30 min. The results obtained revealed that the interaction with the best efficient RB5 removal occurred in acidic to neutral pH. It has been proved by Hashim et al. (2019), that within the acidic to neutral pH, the predominant species is Al(OH)3. The Al(OH)3 species has a huge surface area for efficient dye removal through sweep coagulation and precipitation mechanisms (Kobya et al. 2016). On the other hand, the pH beyond 7 showed a decrease in the RB5 dye removal due to the predominant monomeric species Al(OH)4− . The different forms of charged multimeric hydroxo Al3+ species may be formed as time of treatment increases. These hydroxo cationic complexes are charged efficiently to remove the pollutants by adsorption to produce the charge neutralization, and by enmeshment in a precipitate (Mollah et al. 2001).
Interaction between pH and RB5 concentration
The concentration and the pH interaction in Figure 8(f) impacted the RB5 removal efficiency. The RB5 dye removal efficiency was slightly affected by the RB5 concentration while the former was considerably affected by pH with high removal in the pH range of 4 to 7. The highest RB5 removal efficiency of 70% was obtained with a low RB5 concentration (13.75 ppm) and with a pH range of 4 to 7. The lowest RB5 removal efficiency was obtained with a high RB5 concentration (13.75 ppm) and with a pH of 10. The effect of pH on RB5 dye removal has been previously explained in detail. The slight decrease in RB5 dye removal efficiency at the same current density is due to a high amount of pollutant with the same amount of aluminum cations electrochemically generated in the EC reactor as the current was kept constant at the midpoint.
Optimization
Therefore, the dye removal percentage with predicted factors are current density 1(47.5 mA/cm2), EC time 1(23.75 min), initial concentration − 1(17.5 ppm) and pH − 1(5.5). The maximum predicted dye removal percentage with a desirability value of 1, has reached 96.33%. The reliability of the predicting model was verified through the experimental run with the deduced optimum conditions and the removal efficiency predicted was in good agreement with the experimental results with 94.23% as the RB5 removal efficiency. Therefore, the regression polynomial Equation (14) turned out to accurately model the RB5 dye removal efficiency by EC.
Aluminum dissolution
CONCLUSION
In this study, the EC process was investigated as an eco-friendly technology to treat textile effluents containing RB5 dye. The research focused on the optimization of operating parameters affecting the EC process for the RB5 dye removal by using CCD in RSM from which the regression model was generated to predict EC for the RB5 dye percent removal. Two factors were varied according to the designed experiment and the other two were fixed at zero level, the value that corresponds to midpoints. The operating parameters that were analyzed and optimized are current density, EC time, RB5 dye concentration and pH. The RB5 dye removal increased with increasing current density, increasing EC time and with intermediate or low RB5 dye concentration in weak acid to neutral pH. The optimum operating conditions that were obtained were a pH of 5.5, the current density of 47.5 mA/cm2, an EC time of 23.75 min, and a concentration of 17.5 ppm. The predicted optimum RB5 dye removal was found to be 96.33% and the experimental RB5 dye removal efficiency of 94.23% was obtained with the predicted operating parameters. Thus, the EC process turned out to be an efficient process for RB5 dye removal from wastewater. In addition, the RSM enabled us to optimize the key parameters that affect the EC process where the values from the predictive regression model fit with the experimental values. Therefore, the optimized EC process constitutes an efficient potential technology for textile effluent treatment removal. For further study, the data from EC discontinuous mode should be extrapolated to continuous mode and the techno-economic evaluation should be conducted.
ACKNOWLEDGEMENTS
This work was supported by ISP fund from the SIDA program (Swedish International Development Cooperation Agency) to the University of Rwanda, College of Science and Technology.
DATA AVAILABILITY STATEMENT
All relevant data are included in the paper or its Supplementary Information.
CONFLICT OF INTEREST
The authors declare there is no conflict.